The Palabos Summer School 2020 will take place on 8/10 July 2020 at the University of Geneva - Switzerland and will cover selected topics in Lattice Boltzmann theory and a practical introduction to the open-source Palabos software library.
Participants should have prior exposure to the Lattice Boltzmann Method and Computational Fluid Dynamics. Typical participants could be PhD researchers who are already 1-2 years into their PhD or PostDocs (LB beginners are instead referred to Week 5 of our Mooc). The proposed topics cover:
General framework for modern collision models
Lecturer: Dr. Christophe Coreixas
During the past three decades, numerous collision models have been proposed to improve the numerical stability of lattice Boltzmann methods (LBMs). Roughly speaking, they are distinct with regard to the moment space used for the collision step (raw, central, Hermite, central Hermite or cumulants), the computation of the relaxation time (which is dynamic for entropic and subgrid scale model based LBMs), and the non-equilibrium part of populations (that is recomputed for regularized approaches). This lecture is then dedicated to the presentation of an unified framework that ease: (1) the derivation of relationships between collision models, and more importantly, (2) their implementation in a LB code.
Mesh refinement algorithms
Lecturer: Dr. Orestis Malaspinas
Mesh refinement is of crucial importance for many engineering applications especially for high Reynolds number flows. In the frame of the lattice Boltzmann method, the transitions are particularly sharp because of the cartesian nature of the mesh and must be handled with great care. In this session we will discuss the theoretical aspects of mesh refinement in the lattice Boltzmann method and its implementation in Palabos. Finally we will see how to set up the mesh to perform the simulation of a high Reynolds number flow and then perform an actual simulation.
Curved boundary conditions
Lecturer: Francesco Marson
During this session, we will cover the main techniques to handle complex curved boundary conditions. We will see how to deal with moving boundaries and the different ways to compute the fluid-solid momentum exchange. At the end of the lecture, you will be able to choose the best fitting off-lattice boundary condition for your application and to use it for a fluid-solid interaction problem.
Fluid-structure interaction and simulation of Red Blood Cells
Lecturer: Christos Kotsalos
Fluid-structure/ solid interaction (FSI) concerns how a fluid understands the existence of an immersed body and vice-versa. A consistent and meticulous handling of the FSI is a very critical factor for stable and accurate simulations that deal with both fluid and solid phases. Main focus of the course will be cellular blood flow simulations, i.e., explicit modelling of the trajectories and deformations of blood cells (red blood cells, platelets) inside the blood plasma. More in detail, Palabos will be used for the simulation of the fluid phase (blood plasma) and the immersed boundary method, while an in-house finite element solver will be used for the resolution of deformable bodies physics. By the end of the session, the users should be able to couple Palabos with external solvers and build FSI simulations of complex systems, other than biomedical applications.
Palabos software architecture
Lecturer: Dr. Jonas Lätt
The structure of the Palabos software library is described for two types of target usage, namely the set up new Palabos-based simulations and the development of new algorithms and models in Palabos. The philosophy of Palabos encourages smooth interaction between different aspects of the software, including parallel execution speed, coupling between physical models, execution with different LB collision models or lattice structures, addition of turbulence models, and finally, pre- and post-processing and checkpointing. The class teaches that this interaction is guaranteed by respecting the software model of Palabos, which includes generic collision models (“Dynamics classes”), generic non-local algorithms / couplings (“Data Processors”), and template-driven lattice structures (“Lattice Descriptors”).
Full information can be found here